As many of you know, in Creating Public Value I developed a framework for thinking about strategic management in government that could be reduced to and represented by a ‘strategic triangle.’ (Figure 1)

I argued that it was the important goal of public sector executives to find a ‘fit’ between a conception of the ‘public value’ their organisation sought to produce, and ensuring a continuing flow of resources to the organisation by building ‘legitimacy and support’ in what I call the ‘authorising environment.’ In addition, for any particular conception of public value to be strategically viable, managers would have to be able to build sufficient operational and organisational capabilities to achieve results. What was strategic about this was the idea that we were trying to fit a public sector organisation – led by an executive with control over a bundle of assets and a bundle of capabilities – into a dynamic environment. This environment is, on the one hand, a ‘task environment’ in which we try to orchestrate a portfolio of programs, policies, and procedures to produce public value and, on the other hand, an ‘authorising environment,’ consisting of the citizens, legislators, interest groups, and other parties that tell the manager what particular pieces of the world that executive is authorised and expected to attend to.

When I thought about the standard arguments for performance measurement in the public sector against the backdrop of the strategic triangle, I realised that performance measurement – the process of defining and recognising public value production – fits rather neatly into the strategic triangle and can be used as a highly effective, even essential, tool for strategic management. Take each point in order.

First, at the legitimacy and support point of the triangle, it is, I think, easy to see that one of the best ways to guarantee our legitimacy – to invite support – is to embrace accountability for defining and recognising public value. Indeed, we have a professional and ethical obligation to be accountable for our performance to those who have the power to call us to account and expect us to perform, and this means developing concrete methods of measuring specific aspects of performance.

Second, at the public value point of the triangle, the effort to develop and use performance measures forces us to be much more concrete and explicit about the nature of the public value we are trying to create on behalf of our authorisers. If we cannot make it concrete, we cannot measure it.

Third, at the operational capacity point of the triangle, it is essential that we develop the administrative wherewithal to get the most capacity out of our organisations. This makes it necessary to foster a strong sense of internal accountability, and to give those who work in the organisation guidance so that they can focus on finding the means of improving performance through organisational learning and reflection. These responsibilities depend on the capacity to measure the performance of the organisation concretely and reliably.

So, for a manager standing in the middle of the strategic triangle, it is not hard to see that performance measurement would play an absolutely critical role with respect to handling each corner of this triangle: the definition of public purpose, the mobilisation of support, and the exploitation of operational capacities by driving performance inside the organisation (whether through a rigid structure of accountability or through a softer accountability that emphasises learning).

Performance measurement also plays a key role in bringing the different points of the strategic triangle into alignment. Indeed, managers can use performance measurement to shape their mandate for action. They do this by trying to negotiate the terms of accountability with their overseers in the authorising environment. Given the importance of having a steady and coherent mandate for action in managing their organisations, it might well be worth a public manager’s time and effort to use the challenge of developing performance measures for his or her organisation as a key part of developing their overall strategy.

If we, in our roles as citizen authorisers in democratic countries, were consistent in our expectations of public sector managers, and managers could expect the authorising environment to make sustained and serious efforts to demand performance, then measuring performance would be a very important feature of their task environment. Public managers might feel the need to invest significant effort in negotiations with the political authorising environment to reach some kind of agreement about what they are expected to produce. If public managers were trying to make an agreement with politicians about what they ought to produce and measure, it would be important to them that the politicians not be permitted to change the rules suddenly. If one could get an agreement to stick for a period of time, one would have, in effect, created a policy mandate and, if the authorising environment remained stable in its expectations and demands, managers would at least have some reasonable chance of accomplishing their agreed-upon goals.

Unfortunately, while the political system often says that it wants accountability, and occasionally demands it, it is usually reluctant to negotiate a stable deal about what constitutes public value. There is too much political advantage in retaining the right to change the rules when they want and finding the organisation suddenly accountable for some value that had not previously been treated as important. The reality is that the political system will not allow itself to be negotiated into a binding, long-term agreement. It will reserve the right to change the rules whenever it wants. If we cannot negotiate with the political authorising environment to achieve an agreement about performance measurement, then negotiating in advance will not help us manage organisational performance.

Perhaps, the right thing to do, then, is to appeal to shared virtues of executive responsibility and commitment. To be perfectly candid, I think this is an important part of the duty of a public official and, frankly, I cannot imagine any conscientious public official feeling fully accountable and fully responsible in his job without taking some responsibility for defining whatever it is that he or she is trying to accomplish and making some effort to measure successes and failures. I do not think people can manage public organisations without having a conception of public value that can be measured reliably. Performance measurement is fundamental to the ability to manage organisations. If you do not have such a tool, then I would argue that you are merely presiding over the organisation’s activities. You are not really managing it and you are not creating conditions under which the organisation’s performance could be expected to improve in any reliable way over a period of time. And the reason I think this is so important is that, left to their own devices, organisations tend to accumulate slack, and the only way to move that slack around and get it performing is to embrace a strong form of external accountability in order to animate and drive performance within the organisation.

This is exactly what Bill Bratton did in New York City. When Bratton said, ‘I am going to reduce crime by 10 per cent over the next year,’ he embraced external accountability, set a high standard, and exposed himself and his organisation to the risk of failure. This approach produces a certain amount of energy inside an organisation.

Now, as it turns out, this approach could also be unhelpful in the sense that it might not be a constructive challenge, or in the sense that it might drive the organisation in the wrong direction. However, it is also the case that, if we start running a little bit of voltage through the organisation – not too little, not too much – something begins to happen inside the organisation. The reason I think one should measure performance is that I do not believe one can manage or lead an organisation without some kind of accountability – that is to say, without concrete performance measurement.